Battery arbitrage
Charging a battery when electricity is cheap and discharging when it is expensive, earning the spread.
Battery arbitrage is buy-low, sell-high applied to electricity. A grid-scale battery charges when power is cheap — often midday, when solar floods the market, or overnight — and discharges when power is expensive, typically the evening hours when the sun sets while demand is still high. The profit is the price spread, minus what the battery loses to inefficiency and wear.
What makes it harder than it sounds is that the battery holds only a few hours of energy, so every charge and discharge is a commitment. Discharge too early and you're empty when the real spike arrives; hold too long and the spike never comes. The operator is constantly trading today's known prices against a forecast of the next few hours — and a wrong guess isn't a missed opportunity so much as the position itself.
In ERCOT, arbitrage has grown from a sideline into the main event for storage. Ancillary-service markets, the traditional first stop for batteries, are small and became saturated as storage capacity surged, pushing operators to earn a growing share of revenue from energy price spreads. That shift rewards exactly one skill: forecasting prices and dispatching against them better than the next battery.
For the technical reader
Formally, arbitrage is an optimal-dispatch problem: maximize expected revenue over price paths subject to energy capacity, power limits, round-trip efficiency (typically ~85–90% for lithium-ion), state-of-charge bounds, and cycling/degradation costs, which act as an effective floor on the spread worth capturing. Round-trip efficiency means a battery must sell at a meaningful premium to its charge cost just to break even before degradation.
Real operators run this across markets simultaneously: day-ahead energy positions, real-time deviations, and ancillary capacity awards all compete for the same MW and MWh, making co-optimization under uncertainty — not any single market — the actual problem. Performance is commonly benchmarked against perfect-foresight revenue (the ex-post optimal dispatch), with sophisticated operators capturing a high fraction of it; ERCOT's move to RTC+B changes the structure by having the market itself co-optimize energy and reserves in real time.